A Human-like Solution to Do SLAM (Simulataneous Localization and Mapping) using Fingerprints of Places for Mobile Robots Navigation

Adriana Tapus, University of Southern California
WiSE Postdoctoral Fellow, Robotics Research Lab / Interaction Lab Computer Science Department

The undeniable trend of research in robotics is to endow robots with the capability of understanding the world we are in, thus permitting them to help us  and to be part of our lives. My research is addressing the problem of perception, spatial cognition and topological navigation for a mobile robot. All these elements are needed in order to obtain a robust and reliable framework for navigation. I introduce a new paradigm “fingerprints of places” to solve these problems. Fingerprints of places (i.e. circular list of significant features around the robot) have been proven to be a very promising approach towards effective place characterization and hence environment modeling. I propose an automatic and incremental topological mapping system based on fingerprints of places and a global localizer using Partially Observable Markov Decision Processes (POMDP). The construction of a topological mapping system is combined with localization, both relying on fingerprints of places, in order to perform Simultaneous Localization and Mapping (SLAM). This enables navigation of an autonomous mobile robot in a structured environment without relying on maps given a priori, without using artificial landmarks and by employing a semantic spatial representation that allows a more natural interface between humans and robots. The fingerprint approach, combining the information from all sensors available to the robot, reduces perceptual aliasing and improves the distinctiveness of places. This fingerprint based approach yields a consistent and distinctive representation of the environment and is extensible in that it permits spatial cognition beyond just pure navigation. An exhaustive repertoire of experiments conducted on a mobile robotic platform, has proven this approach to be highly potent. My presentation would detail the complete methodology, experiments conducted, results obtained, problems that have been solved and ongoing research.